306 M. R. Kohlrausch's Theory of the 



must be due to a cause different from the mere loss of the elec- 

 tricity in the surrounding air. 



For the decision of this question, it is in the first place neces- 

 sary to determine accurately the curve according to which the 

 tension diminishes, more especiallj' as this curve is to form the 

 basis of conjectui'es regarding the existence of causes different 

 from those hitherto assigned. Hence our first step is to make 

 the observations necessary for the construction of such a curve. 



If the observations are to be made with purity, care must be 

 taken that the charging of the jar is of momentary duration, and 

 not gradual, so that no uncertainty shall exist as to what occurs 

 during the gradual augmentation of the electricity ; an electro- 

 meter is also necessar}', by means of which the tension may be 

 observed from the moment of charging onwards. The first con- 

 dition is secured by connecting a jar with another, or with a 

 batteiy of jars, already charged ; and the second is secured by 

 the use of the sine-electrometer, applying the method described 

 in vol. Ixxxviii. p. 508 S. of Poggendorff's Annalen. 



In order to interrupt the pursuit of the real question as little 

 as possible, I have deferred to Appendix I. the full description 

 of the mode of observation for such a curve, and moreover for 

 the one marked a in the table which follows, and state here the 

 results merely ; these were obtained with three different kinds 

 of appai'atus, a, b, c. 



a. An ordinary Leyden jar with tinfoil coatings of about five- 

 sixths of a square foot, and in the form of a sugar-glass. 



b. A thick bottle with a narrow, short neck, usually applied 

 to the holding of reagents. The neck was heated, and coated 

 within and without with sealing-wax ; the bottle was then filled 

 up to the sealing-wax with mercury, and was surrounded outside 

 by a laj'er of the same metal, 1 inch thick. To prevent the 

 vessel from floating, it was held fast by means of screws in a 

 cylindrical wooden trough. 



c. A thick rectangular piece of a looking-glass had its metallic 

 coating removed for 2 inches from the edge, and was coated 

 on the other side with tinfoil ; the glass rim on both sides was 

 coated with sealing-wax. During the experiments the plate was 

 laid with the tinfoil upon a sheet of metal, and to the other 

 side, which was connected with the electrometer, electricity was 

 communicated. 



In all three cases, both of the bottles a, b, and the plate c, one 

 of the coatings was carefully connected by a wire with the moist 

 earth of a garden. In the columns marked a, b, c, the results of 

 the observations are stated. In the column headed t, the time 

 after charging is expressed in seconds. The column L^ contains 

 the magnitude of the charge as given by the sine-electrometer. 



